Shiqi Wang

Title: Shiqi Wang: Innovator in Machine Learning Robustness

Introduction

Shiqi Wang is a prominent inventor based in Brooklyn, NY (US). She has made significant contributions to the field of machine learning, particularly in enhancing the robustness of models against adversarial examples. Her innovative approach has garnered attention in the tech community.

Latest Patents

Shiqi Wang holds a patent titled "Adaptive robustness certification against adversarial examples." This patent focuses on adaptive verifiable training, which enables the creation of machine learning models that are robust with respect to multiple robustness criteria. The training method exploits inherent inter-class similarities within input data and enforces multiple robustness criteria based on this information. By utilizing pairwise class similarity, her approach improves the performance of robust models by adjusting robustness constraints for similar and dissimilar classes.

Career Highlights

Shiqi Wang is currently employed at International Business Machines Corporation (IBM), where she continues to push the boundaries of machine learning research. Her work is instrumental in developing models that can withstand adversarial attacks, which is crucial for the reliability of AI systems.

Collaborations

Shiqi has collaborated with notable colleagues, including Kevin Eykholt and Taesung Lee. These partnerships have further enriched her research and contributed to advancements in the field.

Conclusion

Shiqi Wang's innovative work in machine learning robustness exemplifies her commitment to advancing technology. Her patent and ongoing research at IBM highlight her role as a key player in the development of secure AI systems.